The goal of this project is to develop improved methods for diagnosing and treating the ocular complications of the acquired immunodeficiency syndrome (AIDS). This project encompasses clinical trials evaluating new diagnostic and therapeutic approaches for patients with AIDS related eye disorders as well as natural history studies of patients with Cytomegalovirus (CMV) retinitis. Cytomegalovirus (CMV) retinitis is the most common intraocular infection in patients with AIDS and tends to occur after CD4+ cell counts decrease to less than 50 cells/mul. Although anti-CMV therapy with ganciclovir, foscarnet sodium, or cidofovir initially leads to inactivation of the retinitis, the disease progresses in almost all patients despite continued therapy because of inadequate control of the replicating virus. Recent studies have shown that treatment with highly active combination antiretroviral therapy, consisting of protease inhibitors and nucleoside analogs, has lead to decreased human immunodeficiency virus (HIV) loads and increased CD4+ counts. We have been investigating the effect of immune restoration following combination anti-HIV therapy on CMV retinitis. This last year we reported 4 patients with AIDS and increases in CD4+ cell counts induced by highly active combination antiretroviral chemotherapy who had persistently inactive CMV retinitis despite no specific anti-CMV medications. The NEI is now conducting a clinical trial to determine whether elevated CD4+ cell counts resulting from medications against HIV are effective in controlling CMV retinitis in 12 patients who will discontinue specific anti-CMV medications. The primary endpoint of the study will be progression of CMV retinitis. Secondary endpoints will include the occurrence of extraocular CMV disease, morbidity, mortality, virologic and immunologic data, and HIV burden. Enrollment into this study should be completed within the next year. The NEI has been interested in determining which clinical and laboratory parameters predict the development of CMV retinitis. We reviewed the medical records of 234 HIV seropositive patients seen at the NEI between 1980 and 1993. Ninety-three (39.7%) of the patients developed CMV retinitis. The sensitivity and specificity of a CD4+ cell count less than or equal to 50 cells/mul for CMV retinitis was 96.9% and 48.7%, respectively. Although the presence of any ocular symptoms was common, the specificity of ocular complaints for CMV retinitis was only 29.1%. The complaint of new floaters appeared to be a fairly specific predictor for CMV retinitis, with an odds ratio of 3.93 (95% confidence interval 1.86, 8.61, p=0.0004). Our analysis showed that if routine ophthalmologic screening is started when patients develop new floaters, or a CD4+ cell count of 50 cells/mul or less, few cases of CMV retinitis would be missed. Cases of AIDS and CMV retinitis are increasing in less developed countries where CD4+ cell counts are not available. Patient education to improve the recognition of new ocular symptoms like floaters could lead to earlier diagnosis and treatment of CMV retinitis. Finally, a number of medications used to treat AIDS have ocular side effects. We first reported that the antiretroviral agent didanosine caused retinal toxicity. We have been using the electro-oculogram (EOG), which measures the electrical signal from the retina, to monitor the potentially toxic effect of didanosine on the retina in HIV-infected children. Over the past year we discovered that didanosine therapy was associated with a significant decrease in both dark trough and light peak components of the EOG. We found that standard reports of the EOG missed these abnormalities, suggesting that a detailed analysis of the electrical recording is needed to accurately describe the diminished retinal function that can be associated with didanosine therapy.